# FUSEye: Training-Light Fisheye Detection with Overlapping Views and Zero-Initialized Adapters

arXiv: https://arxiv.org/abs/2610.02799 · HTML: https://mlsanity.com/paper/2610.02799/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 37 references that are in the map.
- First posted on arXiv 2 Oct 2026.

## Builds on (most cited first, 27 of 27)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention](https://mlsanity.com/paper/2303.16199/) (2023)
- [Learning multiple visual domains with residual adapters](https://mlsanity.com/paper/1705.08045/) (2017)
- [DETRs Beat YOLOs on Real-time Object Detection](https://mlsanity.com/paper/2304.08069/) (2023)
- [YOLOv10: Real-Time End-to-End Object Detection](https://mlsanity.com/paper/2405.14458/) (2024)
- [YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information](https://mlsanity.com/paper/2402.13616/) (2024)
- [YOLOv11: An Overview of the Key Architectural Enhancements](https://mlsanity.com/paper/2410.17725/) (2024)
- [Soft-NMS -- Improving Object Detection With One Line of Code](https://mlsanity.com/paper/1704.04503/) (2017)
- [WoodScape: A multi-task, multi-camera fisheye dataset for autonomous driving](https://mlsanity.com/paper/1905.01489/) (2019)
- [Learning non-maximum suppression](https://mlsanity.com/paper/1705.02950/) (2017)
- [Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection](https://mlsanity.com/paper/2202.06934/) (2022)
- [Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models](https://mlsanity.com/paper/2606.03748/) (2026)
- [AutoFocus: Efficient Multi-Scale Inference](https://mlsanity.com/paper/1812.01600/) (2018)
- [Generalized Object Detection on Fisheye Cameras for Autonomous Driving: Dataset, Representations and Baseline](https://mlsanity.com/paper/2012.02124/) (2020)
- [Surround-view Fisheye Camera Perception for Automated Driving: Overview, Survey and Challenges](https://mlsanity.com/paper/2205.13281/) (2022)
- [Parameter-Efficient Fine-Tuning for Pre-Trained Vision Models: A Survey and Benchmark](https://mlsanity.com/paper/2402.02242/) (2024)
- [FishEye8K: A Benchmark and Dataset for Fisheye Camera Object Detection](https://mlsanity.com/paper/2305.17449/) (2023)
- [Adaptable Deformable Convolutions for Semantic Segmentation of Fisheye Images in Autonomous Driving Systems](https://mlsanity.com/paper/2102.10191/) (2021)
- [FishFormer: Annulus Slicing-based Transformer for Fisheye Rectification with Efficacy Domain Exploration](https://mlsanity.com/paper/2207.01925/) (2022)
- [Convolution kernel adaptation to calibrated fisheye](https://mlsanity.com/paper/2402.01456/) (2024)
- [SpotPatch: Parameter-Efficient Transfer Learning for Mobile Object Detection](https://mlsanity.com/paper/2101.01260/) (2021)
- [Woodscape Fisheye Object Detection for Autonomous Driving -- CVPR 2022 OmniCV Workshop Challenge](https://mlsanity.com/paper/2206.12912/) (2022)
- [FishRecGAN: An End to End GAN Based Network for Fisheye Rectification and Calibration](https://mlsanity.com/paper/2305.05222/) (2023)
- [FisheyeDetNet: 360deg Surround view Fisheye Camera based Object Detection System for Autonomous Driving](https://mlsanity.com/paper/2404.13443/) (2024)
- [Deformable Convolution Based Road Scene Semantic Segmentation of Fisheye Images in Autonomous Driving](https://mlsanity.com/paper/2407.16647/) (2024)
- [Streaming Object Detection on Fisheye Cameras for Automatic Parking](https://mlsanity.com/paper/2305.14713/) (2023)

## Abstract (arXiv)

Fisheye cameras give mobile robots a single-sensor, low-cost view of their surroundings, yet the COCO-pretrained detectors that practitioners routinely reuse fail on them: strong radial distortion warps local image structure, while boundary compression shrinks objects to near-invisible sizes. Full fine-tuning closes much of the gap but requires abundant fisheye labels and compute. We present FUSEye, a training-light framework that turns a frozen-backbone COCO-pretrained extra-large YOLO26 detector (YOLO26-x) into a fisheye detector. FUSEye adds roughly 227k new parameters while updating the inserted modules and the pretrained detection head. It addresses the transfer gap at three causally linked levels. At the input level, overlapping grid view generation and box remapping (GridViews) enlarge compressed boundary regions. At the feature level, zero-initialized residual adapters (Z-Adapters) correct distortion-induced feature misalignment. At the decision level, learned cross-projection agreement fusion (AgreeFusion) promotes low-confidence detections only when they are supported by consistent evidence across multiple views. On the WoodScape surround-view fisheye benchmark, FUSEye raises YOLO26-x from 0.148 to 0.266 mAP50 and retains 84.3% fully fine-tuned accuracy. Moreover, randomly using only 25% of the labeled training images, FUSEye achieves 0.2597 mAP50, retaining 97.6% of its full-label performance. FUSEye also consistently improves YOLOv8-11 detectors, showing that the recipe is architecture-agnostic. Source code will be available at https://github.com/Su-wenya/FUSEye.